Standard Contract of Supply (dataset)

Abstract

This dataset contains two machine translations of the same English contract of supply with some plain language features made using DeepL Pro in March 2022 and March 2023. The contract was used in my practice as a legal translator trainer. It proved easy to understand but somewhat difficult to translate into Polish, where plain legal language is much less developed than in English. The conclusion from the analysis of renditions of plain language features is that English pronoun you in this context constituted a major source of difficulty and was translated into Polish in many ways. Both translations are also uneven in terms of register (formal legal language vs language of instructions) or forms of address (degree of politeness), which would make post-editing them demanding. Other problems that a post-editor would face include terminological inconsistency and increased number of nominalizations compared to source text. Exponents of modality are rendered mostly correctly. The active voice or short sentences are unproblematic features for MT.

Dane badawcze powiązane z artykułem Is Plain Legal Language Easy to Translate? Plain English features in machine translation of a contract into Polish: https://doi.org/10.55019/plr.2023.1.27-44